with Spencer Greenberg
the podcast about ideas that matter

Episode 099: Volunteering to be infected with COVID — for science! (with Josh Morrison)

Enjoying the episode? Want to listen later? Subscribe on any of these apps or stores to be notified when we release new episodes:

April 7, 2022

How can we develop vaccines more quickly? What kinds of study designs are used (or could be used) during vaccine development? In pandemic situations, we need to roll out vaccines quickly; but even if we can develop and test a vaccine quickly and thoroughly, how confident can we be that there won't be long-term risks? Between ethics and pragmatics, which facet should communicators emphasize when trying to convince organizations and institutions to adopt certain vaccine development strategies? Informed consent is, of course, a hugely important requirement for using human volunteers in challenge trials; so if some people are informed, eager, and willing to volunteer their health and safety for such trials in order to aid vaccine development, then why aren't they being used more (if at all)? Since IRBs are often "all brakes and no gas", could they be given powers to accelerate research in addition to their current powers to slow or halt research? How can bioethics reviews be improved?

Josh Morrison is a serial social entrepreneur and aspiring effective altruist who has founded 1Day Sooner, Waitlist Zero, and the Rikers Debate Project. His work — in the fields of clinical trials, living organ donation, and criminal justice reform — focuses on empowering particular identity groups (research participants, kidney donors, and incarcerated people) to improve decision-making processes so as to achieve impactful policy. His projects have been covered globally in outlets such as the New York Times, CNN, the Financial Times, the BBC, Der Spiegel, Asahi Shimbun, and the South China Morning Post. His writing has been published in the Washington Post, the American Journal of Bioethics, Clinical Infectious Disease, Vaccine, Risk Analysis, Vox, STAT News, and BMJ Opinion. You can email him at

JOSH CASTLE: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you've joined us today. In this episode, Spencer speaks with Josh Morrison about vaccine development, challenge studies, and ethical consent systems.

SPENCER: Josh, welcome!

JOSH MORRISON: Hi Spencer, glad to be here.

SPENCER: One thing that's been on a lot of people's minds lately is, how can we develop vaccines more quickly? And in particular, can we make things like a COVID vaccine or Coronavirus vaccine faster in the future, should there ever be another pandemic or should COVID mutate? I know you've done a lot of thinking around this, so do you want to get us teed up with that topic?

JOSH MORRISON: Yeah. My work was my work the last two years, as working on something called human challenge studies. Those are studies where people are deliberately exposed to an infection as an experiment in a controlled setting and then quarantine. The work started with advocating for COVID challenge studies. There are COVID challenge studies, there's one that just reported results, there's another at Oxford that's in progress but they really weren't done very quickly enough to make a real difference for the first generation of vaccines. Basically, the appeal is that you can learn more about a disease if you studied up close from the moment of exposure. Instead of requiring a study of 10,000/20,000 people to learn if something's effective, you can study 100 people and maybe find out in a month if it's effective. That's the theoretical promise of challenge studies. They're not perfect, they have limitations. It's not the same as natural infection, but it can teach a lot and we use it for other diseases. 1Day Sooner, the group I run, is aimed at representing people who want to be in those studies and trying to get them used well for coronaviruses and other types of research to accelerate vaccine development and learning about diseases.

SPENCER: So let's break it down for a moment. Suppose that you had been in charge of running a challenge trial as soon as we knew COVID was going to be a global pandemic. What kind of trial would you actually have wanted to design?

JOSH MORRISON: The first thing you probably want to do is, you want to know what is this like in seronegative people, that is people who have not been infected yet. You'd want to have your study to start with, to find the right infectious dose to see what's going to infect a certain percentage of people who haven't been infected before. But I think, ideally, what would have happened is, at the same time as you were trying to figure that out, you could have tested some of the early vaccines. Ideally, also, you could have tested some people who'd been infected with COVID and also, in a really perfect world, people who've been infected with SARS one. So what that kind of tells you is, in a few months, you would have been able to get an early signal of number one, "Are these early vaccines effective or not?" We now know these vaccines are effective and the challenge model likely would have helped find that and that would have been really useful to know. And number two, "When they're effective, why is that? And when they're not effective, what's the difference?" That's what's called correlative protection. Because what you want to do is you want to be able to say, "Okay, if another vaccine hits this mark, if it achieves these goals, then we know that that immune response that it achieves is going to correlate with production." It's going to mean that you're not going to be infected, at least in a challenge model. That's something that the really close observation of the challenge phase could help with and could have helped us get more vaccines in addition to the beginning of vaccines. That's what in an ideal world challenge phase for COVID might look like.

SPENCER: You mentioned a few different study designs, can we go through them one by one? Let's start at the beginning of, you're taking people who have never had COVID, could you walk us through the design of that first trial you talked about?

JOSH MORRISON: Yep. Traditionally, and I think we would, in an ideal world do this differently in a pandemic. But traditionally, when you're trying to "establish" the challenge model, when you're trying to learn what the right doses of the pathogen, the way that you do it is, you'll start with a group of maybe 3 to 5 people. You expose them to the lowest dose of the pathogen that you think is potentially infectious. In the case of COVID or SARS-CoV-2, the virus that causes COVID, I believe that the COVID challenge study that was just completed, they started with what was called 100 TCID50. TCID50 is a unit of virus that explain what that means. You start with 100 of those, and then you take your five people. If you're getting the right rate of infection among those five people, let's say four of them. You're aiming for maybe 60% to 80% usually. I think at COVID, they were aiming at 50 to 70. Maybe a few of them get infected. If that's the case, at that dose, you do what's called dose escalating. You give it to maybe 20 people. Then if you're hitting your target of how infectious it is, that's your dose and you're done. If it's not, it's just expanding, then you dose escalate. You give people usually about 10 times more virus than you are giving them or bacteria than you were giving them before. You do the same thing where you start with five people. I think with COVID they started with three people. Then you see how that goes. Then you dose expand until you found a statistically significant finding of how infectious or how many people you exposed to this challenge strain are actually going to be infected.

SPENCER: What do we learn from that? It's like the minimum effective dose of COVID in order to be effective in sense of someone actually catching COVID but how does that help us?

JOSH MORRISON: It helps us in two main ways. I think the most obvious thing is — if you want to do a future challenge study — now you know that this is the amount that looks reasonably safe and that we think is likely to give us this mystical power that we need. But it's also really useful for understanding the disease itself. Because usually, when you observe an infection, you're observing it after there's some symptom. But what challenge studies allow you to do is two main things. One is you can see it at this really early moment. You can see from the moment of exposure, and you can take all these samples, and you can do all fancy analyses to learn more about the disease. Then also, in addition to when you can see it, you can also watch the course of it under an experimental condition where you know exactly what you gave someone when you gave it to them and you can compare all the different kinds of people in your study in that type of controlled experiment. That's uncommon. That's something you can't do outside of that challenge study.

SPENCER: Let's take the consequences of this further. If we had this trial, what would we have done better in terms of handling COVID?

JOSH MORRISON: Just with that, I think, one thing that comes out of these results is, it teaches us a bit about rapid tests, and about how useful rapid tests are, what percentage of infections they catch, what percentage of viral shedding they catch. That could have been really useful. Because especially early on, it could have really made a difference in thinking, "Well, basically the studies indicate rapid tests can be very helpful from a transmission control perspective." Knowing that early would be good. Another thing we see in this data is how the amount of virus that someone sheds really varies quite a bit person to person. I think that could have given us a hint about super spreading dynamics that we didn't know before at the very beginning of the pandemic. Those are the two most obvious things, looking at these studies results that we could have found. There's other experiments you could do about the effectiveness of masks and things like that. But those are too obvious from what the data ended up being in these COVID challenge studies.

SPENCER: I was gonna ask about that. Because it seems to me, one of the most useful things would have been, to have someone that you know has COVID in a room, have someone that you know doesn't have COVID in that room, and test a whole bunch of different variants of: Are they masked? What kind of masks? How far apart are they? How long are they in the room? Because it seemed like for incredibly long time, people were debating, how well do masks work? What are the dynamics of, does it go through the air? How long do you need to be exposed? And so on. Even today, it feels like there's not complete certainty around that.

JOSH MORRISON: Exactly. I think that's definitely a valuable thing that we could have done with these studies. That's what's called a controlled natural exposure challenge. Out here, they're generally pretty rare. I think that's a shame and it would be useful to do it for this disease. One interesting example of one is Jonathan Van-Tam, who is, I think, Deputy Medical Adviser in the British government, who was one of the British government officials who were involved in approving COVID challenge studies. He actually tried to do an experiment like this for flu. The result was they couldn't really infect people. The design of the experiment was basically they had people playing cards with each other. Now, I think, in retrospect, what we can think is, "Oh, well, that's because it's a respiratory disease, it spreads through the air. It doesn't spread on playing cards and maybe the ventilation they were doing was a lot." So even there, you can see these hints of, maybe if we paid attention a little bit more or thought about these studies a little differently, they could have been really helpful. On the other hand, it's a bit of novel territory and I think it's something where challenges can be really quite valuable.

SPENCER: Now another study design that you mentioned that I'd like to go through with you is around vaccine effectiveness. How would this design work?

JOSH MORRISON: The vaccine efficacy study or vaccine challenge study is really quite a bit like a normal vaccine study or a study to see if a vaccine is effective. Where in a normal study, you take some group of people and you give half of them the vaccine and you give half of them a placebo. Then you allow them to go out and live their lives. Some percentage, maybe 1%, maybe 2%, and 0.1% of the placebo group is going to get infected, and some other percentage of the vaccine group gets infected. Then you compare the two. A challenge study is basically the same thing but instead of having people live their normal lives, you actually expose them to infection. The placebo group should probably have about 70% to 80% of them are going to be infected depending on the results of your infectious dosing study. The vaccine group, if the vaccine is effective, will have much less than that. So you can see much more quickly, basically, as quickly as you can challenge people within a few weeks whether your vaccine is effective. Now I should say there's a caveat about this, which is, they call this a human challenge model by which I mean it's not the same as natural infection. It's a model of natural infection, because of course, naturally, we're not putting droplets of SARS-CoV-2 in your nose. Because of that, it's not perfect, for example, you might not find out things about severe disease, if a vaccine affects severe disease really well. We're hoping that no one gets severe disease in the challenge study, and there's gonna be relatively few people. You won't necessarily find that out with a challenge study. But it's still a good early indicator of effectiveness, particularly because if the vaccine is effective in the challenge study, particularly for respiratory diseases, it seems pretty likely that it's going to be effective for more severe disease and outside of the study as well.

SPENCER: So using this kind of design, we can use a lot fewer study participants and we can get answers faster. Is there any other advantage or is it just about smaller sample size, faster conclusions?

JOSH MORRISON: It's also bettered in more specific data and that's partly a function of it being smaller. It's partly a function of, you have this total experimental control. Because usually, we can take some samples around when you're vaccinated. Regularly, at any point. Once you report symptoms, we can take samples. With a challenge study, we can say, "Okay, before you're vaccinated, we can take nasal swabs and we can try to take some lymphatic fluid and blood and all these different things. And we can do the same thing when you're exposed and then we can find out exactly the moment where it starts replicating and when you're infected. Then we can observe the course of illness for a couple of weeks or however long we need really up close." So that gives us this much more precise snapshot of how the infection works on what's called pathogenesis and the immune response to the infection. That can be really, really helpful.

SPENCER: I think a useful piece of context here is information about how quickly the vaccines were developed. Because if you're thinking about doing a challenge trial to try to speed up how quickly we know the vaccines work, well, that seems much more advantageous if the vaccines can be developed really quickly. Can you give us some explanation of how the development process went?

JOSH MORRISON: Particularly for the mRNA vaccines, the development with COVID was really shockingly fast. Where from the time that (I think in January ) the sequence including the spike protein was sent to scientists at Moderna and BioNTech and elsewhere, to when they actually had the construct that you would put in the vaccine was two days. So it's shockingly quickly. Basically everything from January to November or so, when the vaccines were approved was all about testing those vaccines. First testing, if they were safe in humans and a small group, if they were toxic. Then testing if they generate the immune response. Then testing if they were actually effective and also safe in a broader population. That itself was really fast. That took maybe 10 months. Usually a vaccine, to test it takes more like 10 years. So both sides were very fast. The development side of things, that was really enabled because of pandemic preparedness research that happened ahead of time where the NIH or NIAID, the part of NIH that Fauci runs, commissioned people at the University of Texas Austin, Jason McLellan, to develop this spike protein that was engineered to be locked in place to show the immune system it really clearly and that spike protein was what was used in every Western vaccine besides AstraZeneca. The Oxford team that developed the AstraZeneca vaccine, they had already developed SARS-CoV-1 vaccine candidates. There's a lot of preparation work done ahead of time. There was also preparation that the NIH did in terms of promoting mRNA vaccines as a pandemic platform. There's a lot of work that went into it ahead of time, and then just within a couple of days, they actually had a candidate that ended up working extraordinarily well.

SPENCER: They got this vaccine candidate out amazingly quickly. If we were using challenge trials, how much realistically could it have sped up the whole process? How much sooner could we have vaccines in people's hands than without the challenge trials?

JOSH MORRISON: It depends on a few different questions. One question is, how quickly can you do the challenge studies? Because you have to grow the virus. You have to make sure it's really pure. Then traditionally, that takes several months. Then you do the infectious dosing study which is best case several months. Then you can do your challenge study. I think, in an ideal world, if we were proceeding on an emergency basis, it can make that move more quickly. Maybe if we've been able to do challenge studies starting in April and then instead of having to do the infectious dosing study separately, we had done it simultaneously with the vaccine studies, then you're talking something like early May/early June for having results about efficacy. Now, I should say, that's not quite the same as what we had with the vaccines in November because it also takes a long time to develop the manufacturing capacity. But it seems pretty obvious that if we knew in May or June these vaccines look like they're 90% effective then I think we would have done more to manufacture them. We could have accelerated that manufacture pretty significantly and we could have started vaccinating the highest risk groups much earlier on than we did.

SPENCER: How much earlier on would you guess?

JOSH MORRISON: Again, in a perfect world, if you start in a couple of months and then you get a result in a month or two and let's say it accelerates manufacture significantly but not unrealistic. I think, hopefully you're saving two or three months, I think would be the ideal.

SPENCER: So that's best case scenario. Somewhere between zero and let's say, three months is what you're getting, right? Depending on various parameters and how things go.

JOSH MORRISON: Yeah. For the speeding up the initial vaccines, there's also all these other questions around fractional dosing, and First Doses First, and different things that we never did really answer and challenge studies would have been really helpful to answer even if they were not probably faster for the first vaccines. But I think if challenge studies were used perfectly, and everything went according to best case scenario, then you'd be talking two to three months compared to what we did.

SPENCER: It doesn't seem like a huge amount of time. But on the other hand, the scale of the pandemic and how many people were harmed, that actually could be a ton of good if we got it out that much faster. But can you just mention these other benefits a little bit like fractional dosing?

JOSH MORRISON: I'll start with fractional dosing because there's a paper that just came out recently in the proceedings of National Academy of Science about this question of, "Can we give people half a dose of a vaccine instead of a full dose and have it be just as effective?" Our best guess right now is probably yes. Given how many billions of people around the world have been vaccinated, that could have saved 1000s or even hundreds of 1000s of lives if we had known, let's say at the end of 2020, that you would get just as much effectiveness or close to as much effectiveness from half doses and we could produce more and make those doses go twice as far. Another thing that would have been helpful to have challenge studies for, is an issue called First Doses First. That's basically whether we should have given everyone one dose of the vaccine when it was first available, and then wait three or six months to get the second dose. Or, what we actually did was to give reserve two doses for each person 20 days apart. But then as a result, there are only half as many people protected quickly or getting that initial dose of protection. There was a lot of controversy at the time about this question and it turns out that having the later dose is probably more effective. It probably gives you a better boost than one dose and then 20 days after another dose. So challenge studies, we could have tested the efficacy of that very quickly, pretty low cost earlier on. That's another example where a challenge study might have been helpful. It could also be helpful for not vaccine stuff, but stuff around antivirals. You could test Fluvoxamine or Ivermectin in the challenge study. The advantage of that is just because you have this control population so it's a little bit easier to get your test participants than it would be [inaudible].

SPENCER: You're not just waiting for people to come to the hospital that you can study like, "What if we treat them right away? What if we treat them prophylactically? What if we treat them after two days?" That kind of thing?

JOSH MORRISON: Yeah, exactly.

SPENCER: Now going back to the speeding up of the vaccine being released, one thing that confuses me a little bit about that is, don't we need long term data anyway to be sure that this is safe to roll out to a billion people or 5 billion people?

JOSH MORRISON: What I would say is, what we could have done is provided vaccines on an emergency authorization basis, starting with health care workers or people who were at greater risk where that risk benefit trade-off was different. Then you could gather more data from... If you're giving it to a million people, you'll learn a lot more quickly about some of these safety signals, then in your Phase Three, we're giving it to 30 or so thousand. But also just, either way, in a pandemic, there's a trade-off between speed and the long term safety. With the Phase Three data that we had, also we only had a couple months of data for those vaccines in those studies before they were approved. It was still valuable and it helps exclude a lot of safety signals. But there's going to be a trade-off each way, either way.

SPENCER: Something that surprised me about the way you talk about challenge trials, is that you don't lead with the ethics part, because it seems to me that's the main objection people have. I'm just a little surprised that you didn't just start with, "Here's why this is ethical." Maybe we can dig into that a bit. Am I wrong about that? Is that not what most people's concern us?

JOSH MORRISON: Absolutely. Right. There's an obvious piece, which is... Maybe it's just that I assumed, people hearing about this, they're going to think to themselves, "Wait a minute. Well, you're deliberately infecting someone with a deadly disease. I don't know if that sounds so great." But I think, definitely — we'll talk about the ethics in just a second — but in terms of, why I start with the practicality of it, probably, I think that's what's more important because the practicality is really tricky and difficult. All these different pieces of, well, you have to do this infectious dosing study, you have to produce the virus, it's an imperfect model. Those are really important and really tricky. You can't jump over those to the more fun (in some sense), the more attention-grabbing ethics questions. One part of why that's where my mind starts with. The other part to me is, it just feels very obvious that if you have this disease that's killing 10,000 plus people a day, and you might actually be able to make real progress on that, you might be able to — as the name of organization — you might be able to save a day, or a month or two months. The starting point of that being possible, to me, is just way more important than the risks to the people in the study. Particularly because the people are making a free choice to be in the study. There's a lot of people like myself and the 40,000 or so people who signed up with 1Day sooner who wanted to be in these studies for largely prosocial and altruistic reasons. Any doctor is going to worry in its "First, do no harm," right? So anytime you're deliberately infecting people, that obviously has to meet a very high ethical bar. One aspect of that ethical bar, one thing that happens a lot is, we do challenge studies with malaria, cholera, typhoid, and a lot of very dangerous diseases. But those are diseases that we have really reliable and excellent treatments for. So we'll give someone malaria deliberately but we're quite confident that we can treat that malaria really effectively in a way that we obviously can't with COVID and certainly couldn't have at the beginning of the pandemic. Then on the other hand, there's other diseases like influenza, or RSV (respiratory syncytial virus) which is something that usually causes adults a cold but it kills a lot of children and a lot of immunocompromised people. With those diseases, those are things we don't have effective treatments for. Young, healthy people can die if they get a flu, or if they get RSV. It's just extremely rare. COVID is in between, and it's riskier than challenge studies that we do with diseases for which there aren't treatments. So that and the uncertainties around things like long COVID raise ethical concerns that obviously are a barrier to doing studies really quickly and at a large scale.

SPENCER: Right. I wonder if, just in terms of marketing, if starting with ethics might be more effective. I mean, you know a lot more about this me because you're trying to get these ideas out there. You've had a lot of iteration. But I think that if I were going to talk about these and try to convince people that these are good things, if I was convinced that this is really good for the world, I think I would start saying, "Look, there are tons of people who want to enroll in these trials because they want to do their part to help society and in fact, they're correct. They really would be able to help society a lot if they were allowed but we have all these people blocking them from doing something where they fully understand the risks and they're willing to do it and yet they're not allowed to." I feel like that would be an effective way of talking about the topic. I'm just wondering what you think about that way of presenting it?

JOSH MORRISON: I definitely think that centering the fact that there are people willing to be in these studies — not just willing to — both willing and particularly for COVID, eager to be in these studies, I think that's really important and I hope is persuasive. It's definitely the founding principle of 1Day sooner in a way, it's that, by representing people who want to be in these studies, we can make better ethical decisions about the studies. I definitely think starting there and starting with, "There were 40,000 people in 165 countries that signed up and said, 'Hey, I want to be in one of these studies'". I hope that's persuasive because that's usually what we say. So I probably just done a bad comms job today by starting with the really nerdy science stuff.

SPENCER: [laughs] No, I was just... It intrigued me, that's all.


SPENCER: Let's talk about consent, how does one approach consent for a trial where someone's gonna be infected with a virus that in theory they could die from? Some of them probably will die if you do enough of these?

JOSH MORRISON: Yes. I think I would add to that. One issue is the absolute level of risk and the other issue is uncertainty because we don't know every consequence of being infected, particularly for a new disease. Firstly, I think that we should make these decisions about risk for human research subjects the same way as we make decisions about how people make other risks that they take in their lives. Like driving a car, or the job they do, or things like that. I'm very much a liberal, small-l liberal and my philosophical and political commitments around people should be able to make the choices that affect their lives. My first starting point is, people want to be in a study, they should be able to be in the study. I think the key question is, obviously informed consent, can people understand what the risks are and they rethink. There's actually a very good thing — well, it's not a good thing for the world — but a very good thing for an informed consent perspective for COVID is that it's extremely salient, especially early in the pandemic. You look out your window, everyone knows this is a very risky disease, it's affecting society in this really profound way. So even people who aren't the most statistically numerate are still going to have, I think, the correct visceral reaction to it of, "This is something that has risk to it". Now, the other thing that we did at 1Day sooner is we tried to quantify the risk as precisely as possible and tried to create or we did create this online visualization tool to compare the risk of being in a COVID challenge study to other types of risks and visualize it in various consequences. If you have a good treatment, if you don't have a good treatment, things like that. That's work that particularly David Mannheim and Witold Wiȩcek detected and published a paper in the Journal of Risk Analysis about, was really trying to do our best to quantify what is the actual risk here to help people make better decisions about whether they should be in the studies or not.

SPENCER: That's really cool. Really trying to visualize it to give people a more visceral understanding of the risks. We've all had these experiences, where at the doctor we're given some prescription medicine and it lists 42 things that are horrible that could happen to you. It's completely useless information because you have no idea how likely are they. You generally just ignore them, because you're like, "Every medicine has this long list of terrible things that could happen." Or you sign up for a piece of software and there's a bunch of menus. You click through, "Agree, agree, agree," whatever, you have no idea what that stuff says. I think that it's very easy to make consent be bullshit where the people don't really understand what they're getting into.

JOSH MORRISON: So the question is, what do you do with that? My argument would be and the long term dream of 1Day sooner is to be a union or core, like CRPS of challenge volunteers and be representing research participants collectively. By representing the participants, we can do a better job acting on their interests than the current ethical consent system. Now, like a current ethical consent system would be, there's a lot of experts who are essentially independent of the study team who have to approve what materials are being shown to the people in the study, how you talk about risks, things like that. That's how this problem is, you try to address it by having a lot of smart, diligent people really workshop the language and make really sure that it's very clear. But I agree that it's imperfect and one angle of the solution that we want to explore is this idea of trying to represent participants collectively. Number one, at least the research union can assess and say, "Okay, well, these risks seem reasonable enough for us to recommend." So you can be "Okay, this study was recommended by 1Day sooner." Also that we can actually have this personal relationship with the research participant and try to explain to them as well as possible what the risks are in an easy to understand way which is what we tried to do with the risk minimization stuff we did for COVID.

SPENCER: So are you aggregating a collection of people that are willing to volunteer for challenge trials?

JOSH MORRISON: Yeah, that's the idea. It's easier said than done. By which I mean, we had 40,000 people sign up for COVID challenge studies but the vast majority, like myself, are ineligible. You can be ineligible. I'm ineligible for multiple reasons. The biggest one is I don't live in England. They're only taking people who live in the UK, but also I'm too old as well. So it's a bit of a tricky challenge of how do you bring people together in a useful way and direct them into the studies. But that is something we're working on, we had multiple members of 1Day sooner participate in the COVID challenge studies. We're working with folks on studies for whooping cough and influenza and shigella that are other challenge studies we're hoping to have our volunteers be able to participate and hoping to represent the volunteers in the studies.

SPENCER: Well, you say you're too old? That's interesting because you're not that old, you're not especially old. Tell me about, how are the criteria set? Obviously, with COVID, people who are older have a much more extreme risk. Is the idea to do challenge trials just on younger people? And then if you do that, does that limit the generalizability of the trials?

JOSH MORRISON: The current criteria for the existing challenge studies are that you have to be within 18 to 29 and there's a multifactorial test of different kinds of risk factors and things but they're very, very cautious. There were, I think, something like, 34,000 people who signed up at the start of the screening process for the British challenge studies, specifically people who live in England, in order to get, I think, 34 participants. I might have some numbers off by one or two, but it gives you a sense of...

SPENCER: So only 34 people? I mean, that's tiny.

JOSH MORRISON: Yeah, exactly. Basically, the main thing for COVID is age. COVID works, compared to other potential pandemics, that is a nice feature from a challenge study perspective that the risk does go up exponentially with age. So you can have this younger group where the risks are relatively low, or certainly lower than other decisions about what people do like kidney donation or liver donation. Now, in terms of generalizability, it's really interesting question. We should think of these studies as a model and not as the final word on whether a vaccine works or the final word on how the disease works. Because one issue I mentioned before is this artificial means of infection, we're pipetting it into your nose in the case of COVID and that could be a bit different compared to other things. Another issue is the population. It is questionable how valuable are the studies. I think it depends what it is you want to learn with them. I don't think that challenge studies should be normally a replacement for a Phase Three study, I think it's good to be also getting the natural efficacy data as well.

SPENCER: Could you explain what a Phase Three study is?

JOSH MORRISON: Usually vaccine development, drug development, journalists talk about vaccines... There's Phase One, Two, and Three. Phase One is the first time you're putting into humans. It's like, "Is this really unsafe? Also does it generate immune response?" Phase Two is primarily about the immune response, but also maybe gives you a little bit, potentially, could have some efficacy, depending on the size and the type of disease. Then Phase Three is when you get it to a lot of people. In the COVID studies, it was 30,000 people in a given study and that's when you're finally figuring out, "Okay, is this effective and is it safe in a broader pool of people?" Usually challenge studies don't replace the Phase Three, what they do is they help you, it's called downselect. They help you decide, "Okay, is this vaccine worth spending hundreds of millions of dollars on for Phase Three?" Or they can help you understand better, "What should the dose of the vaccine be if you combine a vaccine with an adjuvant and extra thing that boosts your immune system or immune response? What does that do?" Different things like that but they usually don't completely replace natural infection.

SPENCER: You mentioned Phase One trials, where you're giving healthy people, presumably, a drug that has not been tested in humans? Isn't that a kind of challenge trial or isn't that at least spiritually similar to challenge trial?

JOSH MORRISON: Yeah. It's another example of, we're giving you something and it's not really to benefit you per se. It's just for scientific research. They're not very risky, but they're also not riskless. In that sense, I agree that they're somewhat similar. I think that 1Day Sooner, we want to be getting people who are really working on challenge studies but I think we would love to also be working on Phase One studies as we grow because we think it's a bit of a similar population and similar issues apply.

SPENCER: Is the difference just that in a challenge trial, you're actually infecting the person with the disease and not just with an experimental treatment?

JOSH MORRISON: That's exactly right. In a Phase One study, you're giving people a drug that no human has been given before and that has a certain level of danger. But in a challenge study, you're giving them that drug or that vaccine or drug or treatment and then you're also infecting them with or exposing them to infection.

SPENCER: Right. But from the point of view of the participant, you're being given something that has potential danger and some drugs are really dangerous to take, obviously. Probably on average, the drugs are not as dangerous as the virus that you're being infected with. But still, it does seem like at least it opens the door to challenge trials, and it sets a precedent for them.

JOSH MORRISON: No, I agree. I agree.

SPENCER: You mentioned this British trial that's actually ongoing. What are we hoping to learn from that one?

JOSH MORRISON: There have been two studies that were started in England. One by Imperial College, which was co-run by hVIVO which is a company, a contract research organization which runs challenge studies. They're done exposing people. They're done challenging people. So they reported results on, "What is the infectious dose?" Also on different things about that: "How much do people shed?" Things that can teach us things about Coronavirus. I should mention, this is being done with a "Wuhan strain" of the virus, the original strain. There are some plans or at least some funding to do challenge trials with Delta but the data we have is not about the more recent strains.

SPENCER: Seems behind the times, huh?

JOSH MORRISON: Yeah. It's unfortunate and it goes to this point about we're not moving quickly enough with these studies. It gives up some of the advantage. It is, I think, a common refrain of, even when you do this research, there is this hesitancy or it needs to be all treated with kid gloves and everything needs to be very, very "slow and steady wins the race". As a result, you really do fall behind. The data is a lot less useful than if we had found this out in 2020. So it is a bit frustrating that way. The other things it will... The current study that just concluded, that published some results. They're also publishing... They're planning to publish much more detailed immunological results and explain what the immune response was to the virus, and especially they infected about 50% of the people who were challenged. So it raises the obvious question, "What's different between the people that were infected versus the people who weren't infected?" And you can see that really up close, I've urged them to publish this data as soon as possible even in a raw form. We also urge them to publish the data that they just didn't release much earlier. But they don't really... Since the studies were approved, they don't really listen to us. There's another study that's ongoing at Oxford which still needs participants. So if you live in England, and you think this is interesting, I really recommend doing that. The Oxford study is primarily a reinfection study. Though it's also started with people who have been vaccinated. What that study is about is to again get at this question of "Well, what protects people from COVID versus what responses are not protected?" Because you can see, "Okay, well, if you challenge people who've been infected before and they mount an effective immune response and fight the strain off, what is happening there compared to people that didn't do this?" That's the Oxford study that's happening.

SPENCER: I'd be a little concerned about the sample sizes. I'm not sure which of the two is the 34 people but it seems to me, pretty difficult to figure out things like what was different between these people who had a response and didn't have a response when you're only dealing with 34 people.

JOSH MORRISON: That's a great point. I think part of it is that — and I think also 34 is on the smaller end of infectious testing studies, might be like 40 to 50 — usually, if you're testing a vaccine in a challenge study, you're usually aiming to have about 30 to 40 people who are infected in the placebo arm. Usually, those studies might be about 100 people traditionally.

SPENCER: Well, the vaccine is highly effective. You can tell that with a pretty small sample size, right?

JOSH MORRISON: In this case, we know that that's true. That is the case that you could do it with smaller numbers. But part of the thing to realize is when you're doing a normal Phase Three study where you're letting people get infected naturally, even those numbers of how many infections you actually see in order to approve a drug or vaccine are also actually quite small. So you might really have, I think, the Pfizer study, before they announced results had, I think, 90 infections in the control group.

SPENCER: Like a huge number of people are in the trial, right?


SPENCER: But only because so few people happen to get infected before the trial ended, you're saying, at the end of the day, it's really only 90 data points truly that they're using.

JOSH MORRISON: Exactly. Like you said, because the vaccines were 90% effective against symptomatic disease for that time period, then you can have actually pretty much less than 90 and still have a pretty certain result that you're hitting close to the mark. But one answer to that question is, yes, 34 is a fairly small sample size, it should be a concern. But when you compare it, because you're able to actually get infections and witness them up close, it's a bit less of a concern than it might seem compared to what we get normally which is also still a relatively small number of cases.

SPENCER: What's the current barrier? What are the current barriers to challenge trials being run more often?

JOSH MORRISON: I think that it's this funny thing about collective expectations, which is, if you ask scientists, they'll say, "Yeah, these studies are really useful. We should do them." But the regulators are going to have a tough time with it or the ethicists are gonna have a tough time with it. Then you ask the regulators and they're like, "We approve challenge studies. We like challenge studies. We think they can be useful but people just don't come to us with them." And if you ask the ethicist, they're like, "Yeah, we think it's usually pretty ethical." So I think there's this... Because it is more naturally controversial and because people do worry reputationally, like, "Oh, I'm infecting people with a disease, a especially really scary disease like malaria or something." I do think there's a natural tendency to just to proceed very slowly or not to proceed at all to something. Part of our goal is creating this common knowledge around like, "There's all these different groups that think this is going to be useful. There's a lot of public sentiment in favor of it and that can help accelerate things." That's one reason I think they're underutilized. I think just naturally, we should expect a study where scientists are deliberately infecting people to be underutilized. It's a logical prior to have. But the other thing though is, and this obviously wasn't the case for COVID, but usually the things we want to use challenge studies for are developing vaccines and particularly, developing vaccines to really serious diseases that we don't have vaccines for already and they kill a lot of people. Those diseases disproportionately are the ones that affect people in poor countries so there's less money to be made from them. Just fundamentally, it's also just the same reason that there's not enough vaccine development and not enough vaccine studies in general. There's not enough money in this space for the fact that tuberculosis kills 1.4 million people a year, strep kills half a million people a year, and there's just not enough money in the field for those studies overall. That's also its own trickle down into challenge studies. But the reason challenge studies uniquely, I think, are underutilized is partly this common expectation issue.

SPENCER: What about IRB approvals? I know lots of social scientists that complain all the time about how their simple survey that's extremely unlikely to harm anyone took a month to get IRB approval. Or they had to make stupid changes to it that made no sense and people are constantly complaining about their IRB approvals and that's for way less dangerous stuff.

JOSH MORRISON: If you ask ethicists and people at IRB, they'd say, "Look, IRB has approved challenge studies all the time." Obviously, there's a range of ethicists who think they should be improved more or less. But I think, just to be concrete, if you look at the COVID challenge case, those studies were submitted for approval in December of 2020 and they were approved in mid February of 2021. That seems like... I don't think there are any major changes to the study. That seems like an obvious case where that really did slow things down. I think a lot of it... The issue with IRB ease is not fundamentally just, they're going to reject research altogether but that it imposes a lot of friction on and a lot of barriers to doing the research. So even though the IRB would say, "Look, we're happy to approve a Zika challenge study (for example). But it's just this Zika challenge study, that's not quite what we're looking for. Not quite good enough. Come back to us with a revision." It just really adds to the cost of doing the studies. It also means that once you start, let's say, you've started your study and you decide that it's hard to recruit people, you want to increase compensation. That itself could be a multi month process. Or changing all sorts of things. Let's say, similar thing, you're not recruiting enough people, you want to publicize the study in some way. You want to do advertisements. You want to talk to press. Whatever it is. Then you have to go to an IRB or often have to go to an IRB with that request. Again, it just keeps piling on as additional friction. The Zika example I should say, is not a theoretical example. There were Zika challenge studies proposed during the Zika pandemic and a lot of people at the NIH want to do the studies. There was an ethical report that came out that said, "Well, look, Zika challenge studies in general that can be fine. But is this the right time? Are these studies necessary?" Then they weren't done. I think if you read public coverage of it, it really seems like a lot of folks in the NIH wanted to do it. Dr. Fauci did not and they did not happen. I think that didn't really set the tone or set the precedent for what happened with COVID.

SPENCER: Why did Fauci not want them to happen?

JOSH MORRISON: The view that's publicly stated... I think there's two main things that are stated in that report and there's a good STAT news piece that Helen Branswell wrote about this where Fauci's the person who wants to do the study and then Durbin was one of the people explaining, in favor, why it's useful and Fauci was the person quoted as, "Why are they not doing it?" And I think at the time, there were considered two main issues. One issue was, there was a concern that you could be challenged in the Zika challenge study and then afterward, either get pregnant or impregnate someone else. Then they could have birth, the child could then have very serious birth defects. I think a lot of scientists would say that wasn't really very well grounded. But that was definitely a big concern, that uncertainty. The other side of it was, there was a belief that, "Look, there's a lot of Zika out there and we can test a vaccine, naturally through a normal Phase Three study that tests natural infection. So we don't need the challenge studies." And as it turned out, that was incorrect. We actually were able to control Zika through mosquito control. (I think actually like a gene drive which is under discussed.) So there wasn't a lot of Zika and there is, in fact, no approved Zika vaccine. We actually might have one if we had been able to use challenge studies.

SPENCER: If you think about the incentives of an IRB, they essentially, as I understand it, their job is to evaluate the ethics of a study. I can imagine people sitting on that committee, they're getting a lot of stuff that seems pretty normal and then they get a trial saying, "Oh, we're gonna infect people with this virus." They've got to think to themselves that this could get them in deep shit if this doesn't go well. That's their one job. I'm just wondering if there's a big incentive issue here, where if you have a committee that's designed to make things safe, then they're gonna always err on the side of caution. They're not going to be doing the cost benefit and saying, "Hey, this is really, really good for society. It's worth it even though if it goes horribly we're gonna look terrible."

JOSH MORRISON: One way I'd put that is that IRBs, it's like you're driving a car and you only have a brake and don't have a gas pedal. So it can stop a lot. It will stop a lot of research in general just by providing friction and then it will also stop research that's bad by spotting bad research. But it has no way of saying, "Oh. This is important, we should do this, and let's move faster." Or anything like that. I do think structurally, it's problematic for a few reasons. That's one reason. Then as you allude to this rule by committee, it's always going to be, I think, going to tend or it seems like it's going to tend to be more conservative and consensus driven, especially for something like this that might be controversial. And the last thing I'd say is that I think that the risks to participants are not the only risks that the IRB considers. There's also the risk to research in general. Let's say that you do one of these studies and someone dies and then people don't want to participate in medical research overall, or they think that the vaccine is bad. There are risks that aren't about the participants. But I think a lot of the risks are about the participant and so really what you have there is you have people on this committee who are acting as agents on behalf of people who want to be in the study. I think they're very imperfect agents. They're not usually being in this study themselves. They might just be very different, they might have different ideas and beliefs than people want to be in the study. That's part of our theory at 1Day Sooner, we think that having a type of union or liberal or Right to Work union that represents the research participants can do a better job of addressing those ethical concerns than the IRB can necessarily. That we act as better agents on the research participants' behalf than the ethics committee might.


SPENCER: You talked about this idea that maybe if you create a union of people who want to be in challenged trials that that can help speed things up and you also mentioned that maybe you can create more common knowledge that there's actually more support than people might realize from different groups for challenge trials. But what else is in your theory of change of how you get challenge trials to happen a lot more?

JOSH MORRISON: I think part of the idea that this volunteer core model is, we want to be doing more to bring these studies about, and to be more involved in the vaccine development process and trying to answer that and then trying to help. I'll give a couple of examples. One example is with hepatitis C. Some of the people who discovered hepatitis C who won the Nobel Prize for discovering hepatitis C, want to do hepatitis C challenge studies. Hep C is a very, very dangerous disease. So it might seem like, "Wow, that seems really crazy." But actually, we have these really effective treatments now that are 98%+ effective and completely curing Hep C. Then if it fails to heal the first time, you can take it again and it's 90% effective again. The flip side is that, just like we have effective treatments for HIV that don't... About a million people each year still die of AIDS around the world and similarly, about half a million people a year die of hep C. So there's this dire new challenge studies, I've been talking to some of the folks who are working on this and we're helping them to find funding to do a workshop to develop a protocol for these studies. We're planning to help them with working on what the recruitment processes and what the ethics of that should be in general. We're helping with publicity around the studies and trying to be making the case for them. Things like that. Another example is with strep where there's an initiative in Australia, funded partly by the Australian government and by Open Philanthropy, to accelerate strep vaccine development or group A strep is a common term but basically, it's the thing that causes strep throat. That kills about half a million people a year primarily through rheumatic heart disease which it causes. Strep is interesting. It's a bacterium. There's a lot of different strains of it. But the biggest reason we don't have a strep vaccine already is that the FDA essentially banned strep vaccine development from 1979 till 2006. So the growth has been retarded. There was just a challenge model developed in Australia as part of this WHO plan to develop strep vaccines and so we're trying to get funding for that to expand the use of challenge studies in that case and then also expand the strep vaccine development overall by helping fund various phases of the studies. The experience that we had with COVID was, I think, because we were starting from a standing start and weren't experts in challenge studies or in respiratory disease, where we began, we were reluctant to really lead things exactly. We want to be making the case that people wanted to be in these studies, that they were ethical and we were hoping that they would happen naturally otherwise. I think in retrospect, that's a bit of a regret. I think, in the future, we want to be having people representing people who want to be in these studies who really want to bring them about and then our job is to help bring them about. I've mentioned funding. I've mentioned publicity. I've mentioned design. The last thing I'll mention is regulatory science, which is actually pretty interesting. One of the reasons that COVID Challenge stays in happen and that most pharma companies weren't interested, with the exception of Johnson and Johnson, is there wasn't a clear regulatory path of, "If I do these studies and it shows that my vaccine is effective and I get all this cool immune data, what does that do for me? How does that help me get my product authorized?" There really wasn't a good blueprint for that. I also think that there's a lot of uncertainty around producing the challenge pathogen, producing the thing that you'd infect people with, and what the requirements for that would be. I think getting greater certainty about that, and about these regulatory authorization decisions and what trials will go to that, is a way to be really helpful both in a challenge study context and for other vaccine development as well. I'm gonna give a couple examples there. We're doing this work with this panel of different experts and stakeholders, including some folks who run the Office of Vaccine Research and Review at the FDA that's around the next pandemic, and making challenge studies useful for future pandemics. One of the questions there is what we want to do and by we, I mean, humanity, like the NIH, and CEPI which is the Center for Epidemic Preparedness Initiatives, this global multilateral pandemic preparedness entity. There's this plan to develop prototype vaccines. To say, "Okay, well, let's think what are the types of disease and types of infections that could cause a pandemic? And let's develop a vaccine ahead of time that looks like it'll work. Then let's actually do a trial for those vaccines." So we want challenge trials to be a part of it but we also want a clear answer from the FDA and other regulators of like, "Okay, so let's say you do all this, and a pandemic comes around, how are you actually going to make these authorization decisions?" Because that can save some amount of time in the future and it helps you orient your studies to make the service more possible in the present day. Those are some of the other work we do.

SPENCER: I have the perception that there are a lot of people who are against challenge trials. Is that true? And if so, who are these people?

JOSH MORRISON: I think there are a lot of people who are against COVID Challenge trials. Challenge studies which the technical term is controlled human infection studies/controlled human infection models, those are pretty relatively commonly used tools in vaccine development. They were used for the malaria vaccine that was just approved on RTS,S. They were used for typhoid conjugate, a typhoid vaccine that was approved and also for a cholera vaccine that was approved. They've also been really helpful for teaching us all sorts of things about influenza, about yellow fever, and all sorts of different differences. So it's relatively common now and I'd argue that they're underutilized for all the reasons we've talked about. But I think that people who say we should never deliberately infect anyone in a medical experiment for any reason, that I think would be an extreme view. I'm not saying zero people hold that view but that would be really uncommon. For COVID challenge, I would say, it depends where you are. In England, there's pretty universal support for these studies in part just because of the nature of elite dynamics. If the government is saying these studies are good and all the scientific experts, the elites and leadership are saying these studies are good and we should do these studies, it really affects the public's understanding and also what other scientists are saying, at least publicly. In the US, you have the opposite, where Dr. Fauci was pretty, pretty hesitant to do the studies and we decided not to do COVID Challenge studies. I would say, my perception is, it's something like 50-50 (or so) among infectious disease experts and bioethicists, among whether COVID Challenge studies are a good idea.

SPENCER: Well 50-50 is not that high a support, so why are so many people against it?

JOSH MORRISON: I think people want to have this bright line of, "Well, we can do challenge study with a dangerous pathogen but we have to be able to completely have a rescue therapy. We have to be able to completely treat it. So there's really no long term risk at all." I think that not having that bright line with COVID where it's like, young healthy person can get COVID and they can die. Now that we have Paxlovid, maybe things might change, but that drug is scarce and so that raises its own ethical issues. I think that one aspect is there's no guaranteed treatment. Then I think another aspect is the uncertainty. Well, we don't know exactly what long COVID really means. And you might be giving people these long term conditions that we don't know what they're going to look like, how can they consent if they don't understand them? Things like that. Now my view of that question is that I think particularly in a pandemic, really earlier in the pandemic, when you have a disease that now probably more than half the world and almost certainly more than half of Americans have been infected by, it's true that the uncertainty is definitely relevant. But the riskier it is, if most people are going to get this disease, that also means the benefits of treating the disease grow in direct proportion to the risk, so I tend to be less perturbed by the uncertainty question but that's also something people are worried about. The final thing is usefulness of challenge studies question. That comes in a few different flavors. I think that Dr. Fauci's argument has been, "Look, we can get these vaccines by doing regular Phase Three studies through natural infection." Because there was a lot of COVID around in the fall of 2020 and even now. I think the result instead... It's like, "Well, why do these risky studies?" Particularly because they're complicated to set up? They're not perfect. It could give me the wrong answer. The study could say, actually, it doesn't reduce symptomatic infection when it really helps severe disease or something like that. They're not perfect. So why go to the trouble of all that and why risk people's lives if we're going to be able to develop a vaccine normally? And my thought there, my biggest thought there is, I think that the NIH's pandemic preparedness plan for vaccine development worked. It achieved the goals that they wanted to achieve of vaccinating or having available vaccines for every American within 12 to 18 months at the start of the pandemic. That was their goal, they achieved that goal. It's really admirable that they achieved that goal. That was an incredibly impressive thing to do. But it also was really something that was aimed at Americans and not the rest of the world and vaccinating everyone and creating the solutions that are possible for everyone. There's also a satisficing that happens from bureaucratic institutions of, "We'll get it in this amount of time. And maybe there's a thing that might get it a little bit faster, that might make it a little bit better. But we're solving the basic problem and that should be enough." I think that's a bit of an attitude in the context in which these types of studies were looked at in the US.

SPENCER: How do bioethicists tend to feel about it?

JOSH MORRISON: I think, broadly, there's three main camps. There's pro, con, and moderate where in the US moderate is probably the biggest camp. So pro, a lot of people who we work with like Nir Eyal — him with Marc Lipsitch and Peter Smith, wrote the first piece in the journal infectious disease proposing COVID Challenge studies. — He's a population bioethicist. I think he's probably a bit more aggressive than I am, in some ways, but largely for reasons I've explained. Seema Shah is probably the centrist view in American bioethics about COVID challenge. She's at Northwestern and her view is, "Well, we should prepare them. Are they necessary? I don't know. Were these ones good?" She thinks it's a tough call. I think her view is probably the closest to the median bioethicists at the NIH. And folks at the NIH who work on this would be Annette Rid or Christine Grady, who runs bioethics at the NIH and is Fauci's spouse. There's folks like Charles Weijer who are against it altogether. Their view is that it's unethical to do without a rescue therapy, without a really strong therapy. That's some level of the spectrum beliefs.

SPENCER: Those opposed to it, can you steelman their argument? Like let's get into their mind for a second.

JOSH MORRISON: The strongest argument against is that there are these long-tail risks around vaccine confidence and confidence in medical research that a challenge study (maybe), 90% of the time, 99% of the time, nothing goes wrong in any way. They're not misperceived. Doesn't raise these long-tail risks. But 1% of the time, either there's some rumor about how when we take the vaccine, we're actually infecting you with COVID that's driven by challenge studies in some way or just something bad happens in the study. So people think something bad's going to happen when they get vaccinated, or something bad happens in this study, or for whatever reason, people don't want to be in medical research anymore. Medical research is obviously really important. Also, it could jeopardize other challenge studies. If something goes wrong here, if these studies are perceived as unethical politically by the public, that could jeopardize the challenge research in some of the diseases I've mentioned where getting a vaccine for those diseases would save millions of lives ultimately. Or for other diseases, like tuberculosis where we're doing a weakened challenge study (a weakened form of tuberculosis) could be extremely helpful. I think that one steelman argument is that these studies aren't super necessary and they could jeopardize these other important things in these complicated, hard to predict ways. Briefly, I think some of the other arguments that are less compelling to me are the points I've raised about. Like the uncertainty that the risks of being in these studies are uncertain and that you might want to have rescue therapy. I think the other argument that's... There's one argument of, "We can get these results naturally." The other argument, I think, is fairly good is, "This is an imperfect model." The brunt of the imperfect model argument, the argument that these studies are not going to be as good as natural infection, and they could tell us the wrong thing, we could get one of the wrong lessons from them. They're just not very perfect in general. I think the brunt of that is, this type of study is a frontier of our ability to science and medical research. It's our ethical frontier, or scientific, or medical frontier, and trying to assemble that plane in mid-flight during a pandemic really quickly to make them really useful. That's really, really hard. Maybe it's impossible to do well. As a result, we shouldn't be risking people for something that 's not really going to work because it's gonna be too complicated of a challenge. Those are my steelmanning of the arguments against.

SPENCER: I think my biggest concern with challenge trials is that a bunch of them will turn out to be useless and that people who are enrolling in them will overestimate the chance that their contribution is beneficial. I think maybe this is a lot around my view that it's tough to do useful science and a lot of science is not that useful. I'm just curious to hear your response to that.

JOSH MORRISON: On the "Will it be useful?" question, there's a few different elements or ways to think of it but maybe I think that one way to think of it is like... One thing that gave me confidence early in the pandemic is the size of the potential effect was so big that even if you are only having a really small chance at getting the effect, that's really really valuable. Let's say, for example, our organization's called 1Day Sooner. Let's say that there's a 1% chance that these studies are going to be useful enough to develop a vaccine to reduce the day of COVID disease burden.

SPENCER: For the whole world, you're saying? Like one day of total COVID burden that would have been there wouldn't be there anymore because of the trial?

JOSH MORRISON: Yeah. The first thing is just the idea that you can reduce a day or a week or a month, all those options are plausible. Because, I gave an example on fractional dosing, or you can establish correlates of protection that you can approve future vaccines, or you might make your vaccine better. Firstly, you're aiming at a realistic target. It's not crazy to think you could save a month, you could save a week, you save a day. Then even if your expected value is 1% of one day, you're still talking — and I haven't done the math recently of how many people were dying a day — but I think certainly over 5000. So even then you're talking about 50 lives that in expectation the study is saving and that itself on, I think, is pretty conservative assumptions really far outweighs the risk to the volunteers. Even if you're doing 500 or 1000 people's worth of challenge studies, even if lots of studies are useless because lots of researches just doesn't work out the way you want. In general, I still think the expected value still ends up being pretty clearly positive enough to be worth it.

SPENCER: Could you just give us a quick idea of how many lives would we expect to be lost in one of these COVID Challenge trials had it been run?

JOSH MORRISON: I would say the risks to a participant are clearly under 1 in 10,000 chance of dying and 1 in 1000 chance of being hospitalized and they're more likely to be 1/10 of that than to be greater than 1/10 of that. Sorry, I think they're more likely to have a 1 in 100,000 than greater than a 1 in 100,000.

SPENCER: So you might get zero deaths most likely? The most likely scenario in these trials, right?

JOSH MORRISON: Certainly. If you talk to people running the trials, they will say there's no real way that someone's going to die in doing them. I don't think that's the right way of thinking about them. I think we should think of it as dangerous research where we really could do everything right, and someone could die and that would be a tragedy, but it wouldn't necessarily mean it wasn't a good idea to do in the first place. So it is still... The way I think of it is that I have confidence that it's under 1 in 10,000. Because you're taking this young healthy population, you're giving them the best medical treatment available. All these different things. Most likely, I have to look at our calculator again, but the median result is even if you do a good number of challenge studies, let's say you challenge eight times more people than have been challenged in empirical studies, you do 250 people's worth being in the study. The median case is you're still unlikely to have anyone die and you're unlikely to have anyone have disease that's severe enough to be hospitalized.

SPENCER: Josh, before we wrap up, I'd like to give you a rapid fire round or ask you a whole bunch of quick questions. How's that sound?

JOSH MORRISON: It sounds good. I'm usually slow but I'll do my best.

SPENCER: Universal Coronavirus vaccine development: Coronavirus has been around for a long time. It think they're just part of what we call the common cold or common flu, right? We just all get infected but then don't realize it. What's the chance of developing a general vaccine against them?

JOSH MORRISON: General vaccine against all of them will probably take a while. A vaccine against the things that cause COVID and SARS 1, I think one will be approved in two years.

SPENCER: Awesome. That's exciting.

JOSH MORRISON: I'll talk about my book and then say that we need an operation warp speed 2.0 to develop these vaccines, something we're working on with the Institute for Progress.

SPENCER: Awesome. Do you have some ideas around how bioethics reviews could be improved? Or just give us a teaser on that?

JOSH MORRISON: I think I talked about this a bit. Well, I think two things. One, research unions. Two, I think it would be better to not have IRBs, and just get a license, different licenses to do research. So the same way you have a license to drive a car, or a factory might have a license to pollute a certain amount or license to do a nuclear plant. You should be able to have a license to do different types of medical research.

SPENCER: Presumably you could get pulled if you did unethical things?

JOSH MORRISON: Yep and there's still rules. You still could be liable if you break the rules. But you don't have to apply to someone each time you want to do a given stint, if you're licensed to do it.

SPENCER: One thing I really like about challenge trials is that in some cases, they seem like they can produce much higher quality evidence than other types of trials. Obviously, it depends on the case. But in certain cases, I think that's true. One thing that I've thought a lot about with COVID is there's so many studies done but so many of these studies were low quality cause people were reacting and doing whatever they could do, not what they would ideally do. It seems like you can have hundreds of low quality studies and still not learn very much versus if we had a much smaller number of high quality studies seems like we would learn a lot more. I'm curious if you have any comments on that aspect?

JOSH MORRISON: I'm aware of this problem and I don't have a particularly great response or solution. I think challenge studies, yes, you should, in general, be able to get good data. I think because you're doing something riskier, the scientists are gonna be more careful and they're just naturally gonna have to hit a higher bar. So in that sense, they might be more likely to be higher quality and this is experimentally controlled. I mean, that's the thing that's really exciting about it is that you're not just trying to grab the right people to come into the hospital and hope they say yes to you doing your experimental treatment. So they're better that way.

SPENCER: I know this is a topic that you could do a whole new podcast episode on but what's your quick thoughts about what's wrong with kidney donation in the current environment?

JOSH MORRISON: There's two things. One, donors aren't treated well enough. We should be treating donors as honored public servants, and donors should get (in the US) the best health care money can buy, and they should get stipends, do their follow-up, and they should get reimbursed. That's the most obvious thing in the system. I also think... and this isn't really a policy reform thing. I think it's like vegetarianism. I'm not a vegetarian. I really admire people who are but I think there's this moral resentment that comes in with vegetarians where people are like, "Oh, those vegetarians are always telling me what to do," which I've never experienced. I think it's because people are like, "What does someone else's... the nice thing they do say about my decisions?" Obviously, no one needs to donate a kidney. It's not like no one's obligated to do it. But I think that kidney donation makes people feel weird because it's not right for most people but they wonder, "Well, should I be doing this?" I wish that people just relaxed a bit about that and we're just like "Some people want to donate. They should be supported to donate." I wouldn't be opposed to the government paying people who are donors. I think it'd be better than the system we have now. I think a public support system would be better like health insurance but I think payment would still be better than what we have now.

SPENCER: When you donated your own kidney, it was to a stranger? Am I correct?


SPENCER: And my understanding is that the systems are now set up so it's actually possible you could save even more than one person's life, because you can unlock multiple donations. Do you want to just mention that quickly?

JOSH MORRISON: Yeah. This is something called kidney exchange or paired kidney donation. My first nonprofit job after being a corporate lawyer was working for this guy, Mike Rees at the University of Toledo, who did the first one of these kidney chains. The idea is, let's say that you need a kidney and I want to give it to you but we're not a match. If someone comes in and can donate to you and I donate to the next person who's in that situation and then that person's donor gives further down the line, you create this domino effect. So if I'm willing to donate to anyone, then I can trigger this domino chain which can have multiple people now. Luckily, Good Samaritan donation rates in the US have gone up significantly. It's still relatively rare. It's about 500-600 people in the US do it each year. But when I did it, it was a bit less than 300, maybe even less than that. It's growing exponentially but slowly. So the chains aren't quite as long. It used to be like, you can start a chain of six or eight people or six people on average. Now, it's a bit smaller than that. But it is really cool and it's really valuable. Also the quality of the transplants are better because you can get closer immunological matching.

SPENCER: Did you ever communicate with the person that got your kidney?

JOSH MORRISON: Yeah, we actually became pretty good friends at one point. We both lived in Cambridge. He'd been a semi-pro football player and he was a club promoter before he had kidney failure. He had kidney failure for about eight years before the transplant. He had all these medical issues that got caused by the kidney failure. He was relatively young at the time. He was 40 when he had kidney failure and was 48 during the transplant. Now we talk about once or twice a year, we usually talk about football. I'm a big New England Patriots fan and he hates Tom Brady and likes the Cowboys. That's usually what we talk about these days. But yeah, I got to know him pretty well. Really charming, charismatic guy.

SPENCER: What's the dynamic like? The fact that you may have saved his life?

JOSH MORRISON: Yeah. It's definitely emotionally fraught and it's interesting. My belief about it is, he didn't choose me as his donor. I wanted to help someone. I don't feel like he has any particular obligation to me. That's the way I feel about it. But he does feel obligated and I think that that can create a bit of a burden in the relationship. It makes it a little bit more high stakes for him in a way or a bit more complicated.

SPENCER: What reaction do people have when they find out you donate your kidney to a stranger?

JOSH MORRISON: It's funny like a positive one (usually) or a "Wow!" one. It's funny because for six years, my day job was with an organization. So usually the first time someone met me, they would find out that I'm a kidney donor which... On the one hand, I want people to know about it because it'll make them more likely to donate to strangers. But also, if someone hears this podcast and then a family member ends up needing a kidney transplant, I suspect (or I hope) they'll be more likely to donate because they've just heard someone else who donated who was really happy about it and it was in good health. So I do want people to know but I also find it weird, particularly when people first meet me, that their initial framing for thinking of me is, "Oh, here's this guy, who gave a kidney." But I think people's reaction varies from "Wow, that's really cool and really nice." to "Huh, that's different. Why did you do that?" It's usually somewhere in between.

SPENCER: I remember reading about this guy who gave away most of his money. I think more money than his family wanted him to and then he gave away a kidney as well. And people hated this guy. I mean, obviously, a lot of people thought he was noble and everything but some people wrote him angry letters and stuff. It is fascinating how sometimes people can have an aversive reaction. Have you ever had that experience?

JOSH MORRISON: Not exactly to my face or at least not that I remember. Not in a really like, "Wow, that's horrible. How could you do that?" Although my parents were very against me doing it. They've come around enough. I think it partly goes to how morality works and how moral reasoning works where the idea that someone could do this really self-sacrificing thing that is not something you would do, it raises this question of "Why am I not doing that? And does that mean that I'm bad? Or maybe, they're bad?" And that's the convenient answer if what they're doing is really extreme and crazy and wrong. Because most people aren't giving their kidney and certainly most people aren't giving away 90% of their money. I'm trying to remember the name of that guy because I know who you're talking about. He was a real estate developer and I met him at a dinner that Peter Singer had in Princeton. I met him and his son was actually. Nice guy. Not like all the others said.

SPENCER: [laughs] I see because he gave away his kidney and his money. But maybe with him, there was a different thing going on partly, which is this sense that his obligation to his family should come first, and if you had dependents and you were giving me all your money and your kidney, maybe people would feel like that was actually maybe bad.

JOSH MORRISON: There's a great novel by Marilynne Robinson called 'Gilead'. It's about a Calvinist Reverend in Iowa. It's about parenthood because he has a young son and the main character is dying. His parent was like that. He was also Reverend and he could not stop himself from literally stealing the food off the family's table. It goes to this exact question about your obligations, your dependents and things like that. I certainly don't think it's a good idea. I mean, when I donated my kidney (I'd rather have children now) but I didn't have children, I was very free to do whatever I wanted. And then future family can make their own decisions about me or potential future family. But it's definitely different if you're giving away money that's like your kids college fund or something like that.

SPENCER: Bringing this full circle, I think partly what's going on there is different moral intuitions are conflicting. On the one hand, people say, "Wow, that's so generous to potentially save someone's life or to give them your kidney. Another would say, "Well, this guy has an obligation to his family and he should be devoted to them." That creates this moral conflict. Similarly with challenge trials, I think that same thing is going on, where at least in some cases (I don't know how many) but some, it seems clear that the benefit of the world is definitely on the side of "do the challenge trial". But then it conflicts with this other moral impulse that some people have that you shouldn't harm people in a trial, you shouldn't give people virus on purpose, and so on.

JOSH MORRISON: I agree that they do have these common conflicts. That's also partly the true line between my work with kidney donation and the work with challenge studies. What we're thinking about is that I wish people projected less of their own, "Should I be in a study like this? Am I obligated to be in a study like this?" on to people who do want to do these things like giving their kidney. Just because someone else wants to give a kidney to a stranger or be in a covid challenge study, it's not a moral commentary on "Oh, do you have to do that?" They should be able to do that if they want and the world would be better if we let people do things like that more often, more easily.

SPENCER: Josh, thanks so much for coming on!

JOSH MORRISON: Yeah. Really appreciate you having me!


JOSH CASTLE: A listener asks, what is your take on the progress and pitfalls of the current state of marijuana legalization?

SPENCER: I try to take both sides of an issue before coming to a synthesis. If I think about marijuana, I know that some people find it really beneficial in their lives. I have friends who smoke it and find it really useful. I also think that sometimes it has negative consequences for people. I know people who started using it every day and eventually came to regret that, found that it had negative effects for them whether it's on motivation or just clear thinkingness, maybe made their thinking a little more fuzzy or made them not as ambitious. I also know people who after smoking for a long time, going off of it, they actually found it difficult to go off. I know people say it's not addictive and that habit-forming. But I do know people report that they had some negative side effects going off of it. I also wonder, people are smoking before they drive cars and things like that. I don't think it's as dangerous as alcohol. But I do wonder whether it affects people's reaction time to some extent and could contribute to things like car accidents. So I don't think it's costless to society to have lots and lots of people smoking but I do think a lot of people benefited from it. Also, there's evidence it can be used as anti-nausea medicine for cancer patients and things like that. Overall, I think it's probably good to legalize it. I'm in favor of legalizing it all things considered. In terms of the current state of it, I think it's a little silly how inconsistent the legalization efforts are in the US where you can have it be federally illegal even though it's legal in the state. It just seems very silly to me, it feels like it's probably better for society to legalize it and we should just go all the way and it's stupid how slow and inconsistent it is.




Click here to return to the list of all episodes.


Sign up to receive one helpful idea and one brand-new podcast episode each week!

Contact Us

We'd love to hear from you! To give us your feedback on the podcast, or to tell us about how the ideas from the podcast have impacted you, send us an email at:

Or connect with us on social media: